Bobeck's Career Highlights

Work History & Interesting Projects DPAS Network Control System (DNCS)
This is my personal favorite. It was a Systems Engineering and Technical Assessment (SETA) project under a government funded contract. This project started in December 1988 and proceeded through September/October 1990 and involved a team from HO and MV. Our BL government contact Supervisor was John Waninski and the team consisted of Systems Engineers from HO: Ahmet Arslan (1988-89), Jim Bobeck (1988-90); and Software gurus from MV (1989-90): Dave McDonald, Jeff Gerrish, and Dave Priebe.

Unfortunately, almost all of my electronic versions of documents are no longer in my possession.

I do have one Distributed Restoration performance benchmark here: DNCS Performance Benchmark from DSL (.ps)
And I have one Operations Plan (what the Military would have to re-consider) which contains a small overview section on the algorithm

here:

Operations Plan (.ps)

If you find this information interesting, you may get paper copies from published documents in Linus/ITIS (what used to be the Iternal Technical Document Service - ITDS). You may get paper copies of these by requesting them via Linus/ITIS.

include 1

include 2

This project netted 2 patents for Ahmet, Dave, Jeff, and myself.

Administration and Maintenance Interface (AMI)
AMI is an Element Management System for the control of Message Routers and Digital Video Home Terminals (DVHTs) currently used by CableVision for their Enhanced Pay Per View and Video On Demand services. This was my first shot at OO Analysis. I employed the Rumbaugh OMT methodology. There was no CASE tool, I used paper, pencil, and captured the analysis in Framaker.
Links to:

The last version of my OOA Requirements, may be obtained by following this link.
As an educational experience in OO technology in general and C++ in particular, I used the Microsoft Foundation Classes and the Software Development Kit (1991 vintage) to build an AMI simulator. This progressed in 2 phases. First, in order to understand C++ without getting confused with Windows programming, I built a command line version of the AMI interface. After I was feeling comfortable about C++, I tackled the Windows learning curve. First by trial and error, and later, using the Visual C++ GUI tools. The software design (class diagram) used for these simulators meet the OOA as a set of requirements. The simulators are not complete, and to their level of completeness, have some bugs. To download these simulators, follow the links (TBD):

Command Line Interface AMI Simulator
Windows GUI AMI Simulator amigui.exe

Interdiction Control System (ICS)
The ICS was initiated at a request from PacBell. They wanted to offer Pay Per View and regular Pay Station services on their newly emerging HFC network. Lucent was building an Interdiction Device to sit over the NIU. All PPV and Pay Station channels were encrypted at the Interdiction Device and were decripted for a period of time as controlled by the ICS. The ICS took it's marching orders from ASOS regarding scheduling when interdiction was to be on/off.

This was my second shot at OOA. This time, I used the Rational Rose CASE tool.
Links to:

Rose Model

Unfortunately, PacBell was unwilling to sign the contract, so after 6 months of work by a team of Lucent employees, we stopped.

Cable Modem Research
This project had tremendous potential. We investigated and pushed for the ADAPt+ protocol to deliver IP and ATM services over HFC to devices at home, behind the cable modem. We preceded the MCNS specification with ours. In my opinion, they pinched many of our concepts. I established that using MCNS atomic components, our ADAPt+ protocol could be constructed, although somewhat less efficiently than we specified it.
Links to:

Schematic Architecture

This project resulted in several patent applications, two with my name. One was for the ADAPt+ control message architecture and messages. One was for a procedure to fall-back to the low speed downstream telephone link in one-way cable plant under a cable failure (under the philosophy: 33.6 Kb/s is not at good as 10 Mb/s, but it's infinitely better than 0 Mb/s).

Internet Telephony Research
This project delt with delivering Voice over IP over HFC. I learned H.323 during this project, and after presenting my tutorial version of it to two customers, we had them chomping at the bit, wanting to deploy it. I left this project before we delivered, but I understand that this is still a viable Lucent opportunity.
Links to:

Problem Statement
Schematic Architecture
Detailed Architecture

It is my undertanding that the current architecture being implemented differs from that at the links. However, the links do point out how H.323 fits into delivering basic as well as enhanced telephony services.

Miscellaneous:
Training At AT&T/Lucent Bell Labs
CV (long long list, precedes AT&T/Lucent employment)